Availability for enzyme-catalyzed oxidation of cholesterol in mixed monolayers containing both phosphatidylcholine and sphingomyelin

Chemistry and Physics of

ELSEVIERSCIENCE IRELAND

Chemistry and Physics of Lipids 71 (1994) 73--81

LIPID$

Availability for enzyme-catalyzed oxidation of cholesterol in mixed monolayers containing both phosphatidylcholine and sphingomyelin Peter Mattjus, J. Peter Slotte* Department of Biochemistry and Pharmacy, BioCity, Abo Akademi University, P.O. Box 66, 20520 Turku, Finland

(Received 28 October 1993; revision received 21 January 1994; accepted 3 February 1994)

Abstract

In this study we have examined the interaction between cholesterol and phospholipids in monolayers using cholesterol oxidase (Streptomyces cinnamomeus) as a probe. Monolayers containing cholesterol and phospholipids in different molar ratios were exposed to cholesterol oxidase at a lateral surface pressure of 20 mN/m (at 30°C). The rate of cholesterol oxidation by cholesterol oxidase was faster in a monolayer consisting of a mono-unsaturated phospholipid (either l-stearoyl-2-oleoyl-sn-glycero-3-phosphocholine (SOPC) or N-oleoyl-sphingomyelin (O-SPM)) and cholesterol than it was in a monolayer of a saturated phospholipid (either 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC) or N-stearoyl-sphingomyelin (S-SPM)) and cholesterol. This suggests that the susceptibility of cholesterol to oxidation by cholesterol oxidase was markedly affected by the phospholipid acyl chain composition. In addition, cholesterol was oxidized more readily in a phosphatidylcholine-containing monolayer as compared with a sphingomyelin monolayer (at a similar degree of acyl chain saturation). The average rate of oxidation, as a function of the cholesterol/phospholipid (C/PL) molar ratio in a binary monolayer (with cholesterol and one phospholipid class), was linear except for one discontinuity, at 1:1 for phosphatidylcholine monolayers (either SOPC or DSPC) and at 2:1 for sphingomyelin monolayers (O-SPM or S-SPM). We interpret these discontinuities as indicating the stoichiometry at which cholesterol can exist dispersed in the monolayer without lateral segregation into cholesterol-rich clusters. Next, ternary monolayers were examined (with cholesterol and one phosphatidylcholine and one sphingomyelin species). In ternary monolayers where sphingomyelin dominated (67 mol% sphingomyelin and 33 mol% phosphatidylcholine plus cholesterol) the stoichiometry was 2:1 irrespective of the acyl chain composition of the phospholipid species. In ternary monolayers where phosphatidylcholine dominated (67 mol% phosphatidylcholine and 33 mol% sphingomyelin) the stoichiometry was 1:1 for monolayers containing SOPC/S-SPM, SOPC/O-SPM and DSPC/S-SPM but 2:1 for the system DSPC/O-SPM. A further titration of the DSPC/O-SPM system showed that the 2:1 stoichiometry changed to a 1:1 stoichiometry only when the O-SPM content decreased below 5 mol%. The observation that such a small amount * Corresponding author. Abbreviations: COase, cholesterol oxidase; C/PL, cholesterol/phospholipid molar ratio; DSPC, 1,2-distearoyl-snglycero-3-phosphocholine; O-SPM, N-oleoyl-sphingomyelin; PC, phosphatidylcholine; SOPC, l-stearoyi-2-oleoyl-sn-glycero-3-phosphocholine; S-SPM, N-stearoyl-sphingomyelin.

1. Introduction

Cholesterol is a ubiquitous membrane-active sterol, which modulates the fluidity and behavior o f the phospholipid acyl chains [1]. The interac-

0009-3084/94/$07.00 © 1994 Elsevier Science Ireland Ltd. All rights reserved. SSDI 0009-3084(94)02306-P

74

P. Mattjus, J.P. Slotte / Chem. Phys. Lipids 71 (1994) 73-81

of O-SPM in a ternary monolayer containing cholesterol and DSPC could influence the susceptibility of cholesterol to oxidation by cholesterol oxidase suggests that co-operative effects, affecting the two-dimensional packing structure of the monolayer, were induced by O-SPM. Key words: Monolayers; Cholesterol oxidase; Cholesterol; Phosphatidylcholine; Sphingomyelin; Lipid interactions

tion between cholesterol and phosphatidylcholine (PC) in model membranes appears to be of an equimolar nature, and the components are miscible (and thermodynamically stable) at the cholesterol-to-phospholipid (C/PL) molar ratio of 1:1 [2-4]. If the C/PL molar ratio is less than 1, clusters of free phospholipid coexist with 1:1 cholesterol/phospholipid domains [51. If the C/PL molar ratio exceeds 1:1, on the other hand, free cholesterol clusters coexist with cholesterol/phospholipid [2]. It has further been postulated that at low C/PL ratios, clusters or domains with a calculated stoichiometry of 1:2 could also exist in model membranes [6]. At low cholesterol concentrations (below 0.2 cholesterol mole fractions), spin label studies of cholesterol/PC mixtures suggest that a fluid pha~e containing cholesterol and PC (cholesterol mole fraction equal to 0.2) exists in addition to a solid PC phase [71. On the other hand, even though it is possible to prepare model membranes with C/PL molar ratios higher than 1:1 [8,9], these are considered to be thermodynamically unstable [4]. Cholesterol appears to have a special affinity for sphingomyelin compared with the other phospholipid classes [10], although contradictory results have been reported [11]. The tighter molecular packing density observed in sphingomyelincholesterol membranes [12] and the significantly slower rate of cholesterol desorption from sphingomyelin membranes compared with membranes of other phospholipid classes [12,13] support the notion of a high affinity interaction between cholesterol and sphingomyelin. When the sterol/phospholipid interaction in monolayers was probed with cholesterol oxidase, it was demonstrated that sphingomyelin appeared to have a greater capacity to shield cholesterol from oxidation by cholesterol oxidase as compared with phosphatidylcholine [14]. These differences in the apparent affinity between cholesterol and sphingomyelin or phospha-

tidylcholine have been detected in binary monolayers, although it is fully realized that biological membranes are much more complex, both in structure and in lipid composition. We have therefore added more complexity to our monolayer membrane system, by including both phosphatidylcholine and sphingomyelin simultaneously together with cholesterol, and with this system examined the interaction between cholesterol and the phospholipids, using cholesterol oxidase as a probe.

2. Experimental procedures 2.1. Materials

Cholesterol (99+%), S-SPM, O-SPM, DSPC and SOPC were obtained from Sigma Chemicals (St. Louis, MO, USA). The phospholipids gave a single spot on thin-layer chromatography plates (Kieselgel 60, Merck) when eluted with chloroform/methanol/acetic acid/water (25:15:4:2 v/v). Cholesterol oxidase (Streptomyces cinnamomeus) was purchased from Calbiochem (San Diego, CA, USA) and was used as delivered. Buffer salts were of pro analysis grade, and the water was purified with reverse osmosis and further purified with a Millipore (Milli-Q UF+) system to better than 15 Mfi/cm. 2.2. Lateral surface pressure versus mean molecular area isotherms

Force-area isotherms of the pure or mixed monolayers were obtained with a KSV 3000 surface barostat (KSV Instruments, Helsinki). The isotherms were recorded at 22°C. Stock solutions of the lipids were made up in hexane/2-propanol and stored at -25°C. The lipid solution was spread on the subface with a Hamilton syringe, and the monolayer was compressed at a barrier speed not

P. Mattjus, J.P. Slotte / Chem. Phys. Lipids 71 (1994) 73-81

exceeding 8 ,/k 2/molecule, min. Data were sampled every 2 s. At least three different runs were performed at each lipid composition, and the error of reproducibility was less than 5 tool%. The compressibility of the monolayer films was calculated as previously described [15]. The extent of cholesterol-induced condensation of phospholipid packing in mixed monolayers was calculated as described in Ref. 16. 2.3. Oxidation of cholesterol in mixed monolayers The assay for monolayer cholesterol oxidation was performed with a Teflon trough. The trough was of a zero-order type [17], although the reaction observed (conversion of 5-cholesten-3B-ol to 4-cholesten-3-one) did not obey zero-order kinetics (since both the substrate and product remained at the air/water interface). The subphase consisted of a Tris-buffer (50 mM Tris-HCl/150 mM NaCI, pH 7.5), and the reaction compartment (28.3 cm 2, 30 ml volume) was magnetically stirred (100 rev/min) and thermostated to 30°C. Constant surface pressure was maintained by computercontrolled compensatory barrier movement throughout the experiment. After the monolayer had stabilized for about 2 min, cholesterol oxidase (23 mU/ml) was added to the reaction compartment. The enzyme was dispensed into 160-#1 aliquots, which were stored at -25°C and were used within 2 h of thawing to 0°C. Three different kinetic measurements were made of each film composition. Calculation of the average oxidation rate was made as described by Slotte [14]. 2.4. Phospholipid assay The concentration of the various phospholipid stock solutions was determined by the method described by Bartlett [18]. 3. Results

3.1. Force-area isotherms of pure and mixed monolayers The phospholipids selected for this study include phosphatidylcholine and sphingomyelin,

75

with variation in their acyl chain composition, so that both saturated (DSPC and S-SPM) and mono-unsaturated species (SOPC and O-SPM) were examined. Force-area isotherms of such pure phospholipids have been reported previously [19,20]. We now report force-area isotherms of these lipids in binary mixed monolayers containing one of these phospholipids and cholesterol (C/PL 1:1), and in ternary mixed monolayers, containing one phosphatidylcholine type (either saturated or mono-unsaturated) and one sphingomyelin type (either saturated or mono-unsaturated) together with cholesterol (C/PL 1:1) on water at 22°C. The binary mixed monolayers containing a phospholipid and cholesterol were expanded when the phospholipid type was mono-unsaturated, and condensed with saturated phospholipids (Fig. 1A-D). The looser lateral packing density in mono-unsaturated phospholipid mixed monolayers was also evident from the calculated compressibility values (Table 1), with the distinction that a mixed O-SPM monolayer was slightly more condensed that the corresponding SOPC mixed monolayer. In ternary mixed monolayers, the concentrations of the phospholipids were chosen so that one phospholipid type was present at 67 mol% of the total phospholipid amount, and the minority species was present at 33 tool%. The C/PL for the force-area isotherms was 1:1. As shown in Fig. I E - H (in which phosphatidylcholines dominated) and in panels I - L (in which sphingomyelins dominated), the force-area isotherms were more expanded when a mono-unsaturated phospholipid species was included. This expected trend was not markedly different, whether sphingomyelin or phosphatidylcholine was the dominating phospholipid. The condensing effect of cholesterol on phospholipid packing in binary and ternary mixed monolayers was larger in monolayers containing mono-unsaturated phospholipids (Table 1), as expected [21]. The condensing effect of cholesterol was fairly low (2.7%) in a DSPC ternary monolayer containing O-SPM (33 mol% of the total phospholipid), whereas it was fairly high (13.5%) in an S-SPM ternary monolayer, containing 33 mol% SOPC (of the total phospholipid content;

76

P. Mattjus, J.P. Slotte / Chem. Phys. Lipids 71 (1994) 73-81 A

7o

B

C

D

60

4O 3O 20

E

10 0

Z

E

I

I

40

60

7O

I0

8

I 4-O

E

0 6

I 80

I 40

I 60

I 80

I 40

G

F

I 60

80

i 60

i 80

H

6O

40 3O o 0 ~c ol

20 10 0

ID O

I

I

40

60

i 80

l_ i 60

I 40

I 80

; 40

I 60

~ 80

i 40

70

K

J

I

60

L

50

3O 2O 10 0 ' 4O

6

'o

i

80

40

i 60

i 80

h 40

i 60

I 80

i 40

i 60

J 80

Mean m o l e c u l a r area (~,2)

Fig. 1. Force-areaisotherms of mixed monolayerscontaining equimolar amounts of cholesterol and phospholipids. The isotherms werecollectedat 22°C on pure water. The following mixedmonolayerswereanalyzed(onlythe phospholipidtype is indicated): DSPC (panel A), SOPC (B), S-SPM (C), O-SPM (D), DSPC (67 moI%)/O-SPM(33 mol%) (E), DSPC/S-SPM iF), SOPC/O-SPM (G), SOPC/S-SPM (H), S-SPM/SOPC (1), S-SPM/DSPC (J), O-SPM/SOPC (K) and O-SPM/DSPC (L).

Table 1). The condensing effect of cholesterol on SPM monolayers containing 33 mol% DSPC was fairly small (Table 1). 3.2. Oxidation o f cholesterol in mixed monolayers

The oxidation of cholesterol in mixed monolayers containing phospholipids [14,16], as well as in pure sterol monolayers [22], results in an expan-

sion of the monolayer area. With the knowledge of the reaction time and the number of cholesterol molecules on the surface of the reaction chamber, one can calculate an average reaction rate for the oxidation. This average oxidation rate versus the cholesterol concentration of the monolayer was obtained for both binary and ternary monolayer systems. The average oxidation rate plotted as a function of the cholesterol/phospholipid molar ratio reveals a stoichiometry in which the slope of the function changes [141. The results with the binary systems are shown in Fig. 2. With both phosphatidylcholines, the discontinuity was at a 1:1 molar ratio, whereas it was at a 2:1 ratio for both sphingomyelins. The break in the line at C/PL 1:1 for phosphatidylcholine, and at C/PL 2:1 for sphingomyelin, is interpreted as indicating the molar ratio below which cholesterol-rich clusters do not exist and hence is a potential measure of cholesterol miscibility in the mixed monolayer. It is noteworthy that cholesterol in sphingomyelincontaining monolayers at a 1:1 molar ratio was not oxidizable (within the chosen conditions), whereas it was readily oxidizable at a 1:1 molar ratio when the monolayer contained phosphatidylcholine (Fig. 2). The average oxidation rate was also observed to be consistently higher in mixed monolayers containing phospholipids with unsaturated acyl chains compared with the situation with saturated acyl chains. 3.3. Oxidation of cholesterol in ternary mixed monolayers

Next, we undertook to study how the presence of two different phospholipids in a ternary monolayer would affect the rate of cholesterol oxidation. Since phosphatidylcholine-containing and sphingomyelin-containing mixed monolayers display a different molar ratio, at which cholesterolrich clusters disappear (see Fig. 2), the question was: which phospholipid (and at what relative concentration) will influence the characteristic stoichiometry in a mixed ternary monolayer? The composition of the monolayers was selected so that two different phospholipid classes were always present (phosphatidylcholine and sphingomyelin), with one dominating at 67 mol% over the

P. Mattjus, J.P. SIotte/ Chem. Phys. Lipids 71 (1994) 73-81

77

Table 1 Force-area isotherm properties of pure and mixed monolayers. Monolayer

A r e a a 2 0 mNtm

A r e a b l : l , 20 mN/m

CondensationC

k d 5 mN/m

composition

(/~ 2)

(A 2)

(%)

10-3 m/mN

Cholesterol DSPC SOPC S-SPM O-SPM

39.5 59.0 68.0 52.0 74.0

-45.4 43.7 43.3 47.1

-7.8 18.7 5.4 17.0

1.3 5.0 19.0 3.0 14.3

67/33 mol% DSPC/O-SPM DSPC/S-SPM SOPC/O-SPM SOPC/S-SPM

55.7 54.5 65.8 61.9

46.3 43.2 46.4 43.2

2.7 8.1 11.9 14.8

7.5 2.7 16.7 17.0

67/33 mol% S-SPM/SOPC S-SPM/DSPC O-SPM/SOPC O-SPM/DSPC

59.7 48.3 64.0 57.2

42.9 44.8 45.4 47.2

13.5 0.1 i 2.3 2.4

9.5 2.7 15.0 12.1

aMean molecular area of the pure compound, or of the sterol-free phospholipid mixture. bMean molecular area of the equimolar sterol/phospholipid mixed monolayer. CCondensation of the molecular packing induced b~' inclusion of cholesterol in the monolayer, given as percentage (see Methods). dCompressibility of the monolayer~ determined at 5 mN/m and given as 10-3 m/mN. 1.0

i

i

i SOPC

i

i

i

/

O "~

0.5 0.5

E i O o O

x

0.0 0

o

~ 1.s

I i

i

0

3

2 i

i

~

1.0

< 0.5

0.0

/ 1

): i

2

i

i

0.0 3 i

1.5

O-SPM

S-SPM L

I

2

1.0

0.5

0.0

3

Cholesterol /

4

1

2

3

4

Phospholipid mole rotio

Fig. 2. The average oxidation rate versus C/PL function. Rates were collected at 30°C on Tris buffer (pH 7.5) with mixed monolayers containing the indicated phospholipids, at the indicated C/PL molar ratios (lateral surface pressure 20 raN/m). Values are averages from three different experiments at each molar ratio.

other (at 33 tool% relative to the total p h o s p h o l i p id content). The m o n o l a y e r s were further either fully saturated, m i x e d with one s a t u r a t e d phosp h o l i p i d a n d one m o n o - u n s a t u r a t e d p h o s p h o l i p i d , o r fully u n s a t u r a t e d . W h e n the c o n t e n t o f sphingomyelin was 67 m o l % and that o f p h o s p h a t i d y l c h o l i n e 33 tool%, all mixtures showed a s t o i c h i o m e t r y o f 2:1 (C/PL), at which the slope o f the average o x i d a t i o n rate versus C / P L function changed (Table 2). This showed that the presence o f a m i n o r i t y a m o u n t (33 mol%) o f p h o s p h a t i d y l c h o l i n e in the t e r n a r y m o n o l a y e r did n o t influence the typical sphingomyelin stoichiometry as observed in b i n a r y mixed m o n o l a y e r s (e.g. Fig. 2). In mixed m o n o l a y e r s where the p h o s p h a t i d y l c h o l i n e a m o u n t was 67 m o l % a n d the sphingomyelin a m o u n t 33 mol%, a 1:1 stoichiometry was observed with all m o n o l a y e r systems except the one that c o n t a i n e d D S P C a n d O - S P M (Table 2). This m o n o l a y e r h a d a discontinuity at 2:1. T o examine at which relative O - S P M c o n c e n t r a tion the D S P C m o n o l a y e r c h a n g e d from a 1:1

Phys. Lipida 71 (1994) 73-81

78

P. Mattjus, J.P. Slotte / Chem.

Table 2 Oxidation properties of cholesterol/phospholipid mixed monolayers.

average oxidation rate was determined at different C/PL ratios. As shown in Fig. 3A, a very low concentration of O-SPM (somewhere between 0 and 5 tool% of the phospholipids) was needed to change the DSPC stoichiometry from 1:1 to a 2:1 type. When a similar titration was performed with the system SOPC/S-SPM, it was observed that the 1"1 SOPC system changed into a 2:1 type when the concentration of S-SPM increased from 40 to 50 tool% (Fig. 3B).

Monolayer cornposition (67/33 moi%)

Stoichiometry a (C/PL)

Average oxidation rate b ( x 10 n3molec/s)

S-SPM/SOPC S-SPM/DSPC O-SPM/SOPC O-SPM/DSPC

2:1 2:1 2: I 2: !

0 0 0 0

DSPC/O-SPM DSPC/S-SPM SOPC/O-SPM SOPC/S-SPM

2:1 I: 1 I: 1 1:1

0 0.15 0.20 0.22

4. Discussion The lipids of this study were chosen to include phosphatidylcholines and sphingomyelins, containing either saturated or mono-unsaturated acyl chains (at the sn-2 position in PC or the N-linked chain in SPM). The bulky cis A 9 double bond in both O-SPM and SOPC is known to result in a looser lateral packing of the pure monolayer (liquid-disordered or liquid-expanded)ias compared with the situation with fully saturated phospholipids (e.g. DSPC or S-SPM), which gives a liquid-ordered or liquid-condensed force-area isotherm [12,20]. This situation is maintained even in binary (one phospholipid) or ternary (two phospholipids) mixed monolayers containing cho-

aThe cholesteroi/phospholipid molar ratio (C/PL) at which the linear average oxidation rate versus C/PL function displayed a discontinuity. bThe average oxidation rate (1013 molecules oxidized per s) at a C/PL of 1:1 and a lateral surface pressure of 20 mN/m (30°C).

stoichiometry (at 100 mol% DSPC; Fig. 2) into a 2:1 stoichiometry (at 67 mol%; Table 2), the concentration of O-SPM was gradually decreased from 33 tool% towards zero concentration. The effects of this reduction in O-SPM content on the )

I

I

I

I

I

I

I

I

I

1

I

F

B

A

2:1

I

I

I

A A

A

m

2:1

A

v

1:1

1:1

I

I

I

I

I

L

I

I

I

I

I

I

J

I

I

I

0

15

30

4,5

60

7,5

90

105

0

15

30

45

60

75

90

105

Mole % O-SPM /

DSPC

Mole ~ S - S P M /

SOPC

Fig. 3. Oxidation of cholesterol in mixed monolayers containing variable amounts of sphingomyelin relative to phosphatidyicholine. Panel A shows the effect of decreasing the content of O-SPM (in a DSPC matrix) on the stoichiometry at which the average oxidation rate versus C/PL function shows the discontinuity (cf. Fig. 2). Panel B is analogous, except that the content of S-SPM in a SOPC matrix is changed. Values are averages from two different experiments recorded at a lateral surface pressure of 20 mN/m (at 30°C).

P. Mattjus. J.P. Slotte / Chem. Phys. Lipids 71 (1994) 73-81

lesterol (at a C/PL of 1:1; Fig. 1). The looser packing density of monolayers containing monounsaturated phospholipids (both SOPC and OSPM) is thought to be due to the smaller number of van der Waals forces operating at short distances between acyl chains, and between the sterol molecule and acyl chains of neighboring phospholipids [12,23]. In the monolayers with saturated ph6spholipids (e.g. DSPC and S-SPM), more van der Waals interactions are allowed because of the tightly aligned straight phospholipid acyl chains. It has previously been demonstrated that the lateral packing density of sphingomyelins in monolayers is tighter than the corresponding lateral packing density of phosphatidylcholines [121. Our data with binary mixed monolayers are consistent with this observation, since the compressibility value for mixed sphingomyelincontaining monolayers (at 5 mN/m and 1:1 C/PL) was lower than with the phosphatidylcholine system (at similar acyl chain satUration; Table 1). This observation was true both with the saturated phospholipid pair and with the mono-unsaturated pair. Another means of looking at cholesterol/colipid interactions in monolayers is to use cholesterol oxidase as a probe for the oxidation susceptibility of the 3B-hydroxy group of cholesterol at the water/lipid interface [16,19,22,24,25]. The susceptibility of cholesterol for oxidation by cholesterol oxidase in monolayers is related to the degree of exposure of the 3/~-hydroxy group at the water/ lipid interface. In pure cholesterol monolayers, the oxidation rate is mainly a function of the surface concentration of cholesterol [221. In mixed monolayers containing cholesterol and different phospholipids, the oxidation rate is slower compared with the case with pure monolayers (even when correcting for changed surface concentrations of the sterol), apparently because a cholesterol/phospholipid interaction is stronger (more van der Waals forces) than a cholesterol/cholesterol interaction [14,25]. It is not known, however, what effects (if any) the reaction end-product cholestenone has on the rate of cholesterol oxidation. Recent model membrane experiments suggest that cholestenone is distributed randomly in the bilayer plane of DPPC vesicles [261. If this is the

79

case also with complex monolayers, it is possible that the rate of the cholesterol oxidase reaction is enhanced by the formation of the reaction endproduct, since this has a larger mean molecular area and thus would be expected to loosen the packing density of the mixed monolayer. We have previously reported on the oxidation susceptibility of cholesterol in mixed monolayers containing phosphatidylcholines or sphingomyelins with undefined (or mixed) acyl chain compositions [14]. Now we can expand on these previous observations in a monolayer system with acylchain-defined phospholipids. The function of the average oxidation rate versus C/PL ratio is linear, except for one characteristic discontinuity (either at 1:1 or at 2:1 C/PL). This discontinuity has been interpreted as indicating the stoichiometry at which cholesterol-rich clusters form or disappear (depending on whether one looks at the C/PL as increasing or decreasing [14]), and where cholesterol and the phospholipid are miscible. In this study, both sphingomyelin systems gave a 2:1 stoichiometry, whereas both phosphatidylcholine systems gave a 1:1 stoichiometry, consistent with previous observations [14]. The degree of acyl chain saturation did not affect the apparent stoichiometry, at least in binary systems, whereas it appeared to affect the average oxidation rate (the rate was higher in mono-unsaturated monolayers as compared with saturated films). In ternary mixed monolayers containing two different phospholipids, the effect of sphingomyelin (when dominating at 67 mol%) was as dramatic as in mixed monolayers containing sphingomyelin as the only phospholipid type. The average oxidation rate versus C/PL function had always the discontinuity at a 2:1 stoichiometry, and the average oxidation rate at 1:1 C/PL was zero. This shows the effectiveness of sphingomyelin in retarding the oxidation of cholesterol, even when 33 mol% phosphatidylcholine was present (either saturated or mono-unsaturated). With ternary mixed monolayers in which phosphatidylcholine dominated, the situation was different. A 1:1 stoichiometry was observed with SOPC/SPM monolayers and with the DSPC/S-SPM system. The lower amount of sphingomyelin in these systems was not able to completely retard the oxida-

80

tion of cholesterol at the C/PL of 1:1, a finding that is consistent with the 1:1 stoichiometry. However, when the ternary monolayer contained DSPC (67 tool%) and O-SPM, then the sphingomyelin markedly influenced both the stoichiomerry (2:1) and the oxidation susceptibility at 1:1 C/PL (zero oxidation). The effectiveness of OSPM in influencing the DSPC-dominated mixed monolayer was great, since concentrations somewhat less than 5 tool% were enough to influence the behavior of the monolayer. The effect of the low amount of O-SPM on the average oxidation rate of cholesterol in the DSPC mixed monolayer was probably not caused by effects on lateral packing densities, since the mean molecular area of the mixed ternary monolayer (46.3 A 2 at 67 mol% DSPC and 33 mol% OSPM) was only slightly larger than the mean molecular area of the binary DSPC/cholesterol monolayer (45.4 A2; Table 1). It is more likely that O-SPM affected the two-dimensional lateral packing properties of DSPC and cholesterol, so that an apparent 2:1 stoichiometry was formed (as probed by cholesterol oxidase) when the content of OSPM was higher than or equal to 5 tool% of the total phospholipid concentration. In a broad sense, one can speak of co-operativity when a small number of molecules affect the behavior of a much larger number of molecules. In cholesterol/phospholipid model systems, it has been reported that cholesterol can affect the cooperativity of the motion of phospholipid acyl chains [3,27]. In addition, it is also known that cholesterol at very low concentrations can affect the surface viscosity of phosphatidylcholine monolayers [28]. It appears that O-SPM in a saturated matrix containing cholesterol and DSPC could influence the two-dimensional array formation at the interface, a phenomenon that may be related to the fact that a mono-unsaturated molecule is inserted into an otherwise saturated monolayer. In the opposite case, with a mono-unsaturated monolayer (cholesterol and SOPC), a saturated co-lipid (S-SPM) appeared not to induce a co-operative effect on the lateral packing structure, since the system changed from a 2:1 stoichiometry to a 1:1 stoichiometry only when the content of S-SPM increased from 40 to 50 mol% (Fig. 3B).

P. Mattjus, J.P. Slotte / Chem. Phys. Lipids 71 (1994) 73-81

In conclusion, this study has clearly shown that sphingomyelin has a dominant role to play in regulating how the 3/~-hydroxy group of cholesterol is exposed to oxidation by cholesterol oxidase at the water/lipid interface. It is also evident that sphingomyelin under certain conditions can strongly affect the lateral packing structure of mixed monolayers, even when present at very low concentrations. Whether sphingomyelin also has the capacity to affect and regulate the lateral structure of more complex membranes (biological membranes and lipoprotein surfaces) remains to be established. 5. Acknowledgment This study was supported in part by generous grants from the Council of Sciences (Academy of Finland), the Sigrid Juselius Foundation and the Borg Foundation. 6. References i 2 3 4 5 6 7 8 9 10

11 12 13 14 15 16

P.L. Yeagle (1985) Biochim. Biophys. Acta 822, 267-287. H. Lecuyer and D.G. Dervichian (1969) J. Mol. Biol. 45, 39-57. M.C. Phillips and E.G. Finer (1974) Biochim. Biophys. Acta 356, 199-206. J.J. Collins and M.C. Phillips (1982) J. Lipid Res. 23, 291-298. L.R. McLean and M.C. Phillips (1982) Biochemistry 21, 4053-4059. F.T. Presti, R.J. Pace and S.I. Chan (1982) Biochemistry 21, 3831-3835. D.J. Rechtenwald and H.M. McConnell (1981) Biochemistry 20, 4505-4510. R. Freeman and J.B. Finean (1975) Chem. Phys. Lipids 14, 313-320. B. Lundberg (1977) Chem. Phys. Lipids 18, 212-220. R.A. Demel, J.W.C.M. Jansen, P.W.M. van Dijck and L.L.M. van Devnen (1977) Biochim. Biophys. Acta 465, 1-10. Y. Lang¢, J.S. D'Alessandro and D.M. Small (1979) Biochim. Biophys. Acta 556, 388-398. S. Lund-Katz, H.M. Laboda, L.R. McLean and M.C. Phillips (1988) Biochemistry 27, 3416-3423. S. Clejan and R. Bittman (1984) J. Lipid Res. 259, 10823-10826. J.P. Siott¢ (1992) Biochemistry 31, 5472-5477. J.P. Slotte, A.L. Ostman, E.R. Kuman and R. Bittman (1993) Biochemistry 32, 7886-7892. L. Gr6nberg and J.P. Slotte (1990) Biochemistry 29, 3173-3178.

P. Mattjus, J.P. Slotte/ Chem. Phys. Lipids 71 (1994) 73-81

81

17 R. Verger and G.H. de Haas (1973) Chem. Phys. Lipids 10, 127-136. 18 G.R. Bartlett (1959) J. Biol. Chem. 234, 466-468. 19 L. Gr6nberg, Z.-S. Ruan, R. Bittman and J.P. Slotte (1991) Biochemistry 30, 10746-10754. 20 F. Paltauf, H. Hanser and M.C. Phillips (1971) Biochim. Biophys. Acta 249, 539-547. 21 D. Chapman, N.F. Owens, M.C. Phillips and D.A. Walker (1969) Biochim. Biophys. Acta 183, 458-465. 22 J.P. Slotte (1992) Biochim. Biophys. Acta 1123, 326-333.

23 L. Salem (1962) J. Chem. Phys. 37, 2100-2113. 24 J.P. Slotte (1992) Biochim. Biophys. Acta 1124, 23-28. 25 J.P. Siotte and A.L. Ostman (1993) Biochim. Biophys. Acta 1145, 243-249. 26 V. Ben-Yashar and Y. Barenholz (1989) Biochim. Biophys. Acta 985, 271-278. 27 A. Darke, E.G. Finer, A.G. Flock and M.C. Phillips (1971) FEBS Lett. 18, 326-330. 28 R.W. Evans, M.A. Williams and J. Tinoco (1980) Lipids 15, 524-533.